forked from mia/Aegisub
AlignToVideo: Improve rgb->lab precision
This commit is contained in:
parent
26f0713aae
commit
7f4cdaea42
1 changed files with 3 additions and 6 deletions
|
@ -156,12 +156,9 @@ namespace {
|
|||
|
||||
void rgb2lab(unsigned char r, unsigned char g, unsigned char b, double* lab)
|
||||
{
|
||||
double R = static_cast<double>(r) / 255.0;
|
||||
double G = static_cast<double>(g) / 255.0;
|
||||
double B = static_cast<double>(b) / 255.0;
|
||||
double X = 0.412453 * R + 0.357580 * G + 0.180423 * B;
|
||||
double Y = 0.212671 * R + 0.715160 * G + 0.072169 * B;
|
||||
double Z = 0.019334 * R + 0.119193 * G + 0.950227 * B;
|
||||
double X = (0.412453 * r + 0.357580 * g + 0.180423 * b) / 255.0;
|
||||
double Y = (0.212671 * r + 0.715160 * g + 0.072169 * b) / 255.0;
|
||||
double Z = (0.019334 * r + 0.119193 * g + 0.950227 * b) / 255.0;
|
||||
double xr = X / 0.950456, yr = Y / 1.000, zr = Z / 1.088854;
|
||||
|
||||
if (yr > 0.008856) {
|
||||
|
|
Loading…
Reference in a new issue